Electronics > Beginners
Class AB amplifiers: power supply, layout, theory questions
(1/1)
rectumfried:
Howdy,
I recently (begrudgingly) decided it's time to dip my toe in the audio world. I find the science of acoustics fascinating but am definitely turned off by the barriers to entry: primarily the math and the fact that it seems very hard to find a digestable (for the new players!) breakdown of audio amplifier circuits. Anyways,
I bought two chips on a whim: (as an aside, arrow.com is dangerous. Overnight shipping on so many things....)
1. TDA7375V - 2 x 35W dual/quad power amplifier for car radio: https://www.arrow.com/en/products/e-tda7375v/stmicroelectronics
2. LM1875T/LF02:https://www.ti.com/lit/ds/symlink/lm1875.pdf
Figured the car one should be idiot proof, being single rail and what not. After what seemed like forever I finally managed to point-to-point mid air solder this abortion together...I'd take a picture, but frankly it's embarrassing. Anyways the circuit is a carbon copy (component wise at least) to the Fig 5 Double Bridge example they give in the datasheet. For some reason I only get 1 channel though. I am getting an audio signal at the input but nothing at the pins of the IC. Some guidance as to where I should focus my troubleshooting efforts would be appreciated. I wouldn't be surprised if I did the connections wrong. Some things I'm confused about:
-The datasheet talks about how the standby pin needs an external 5ma limited power supply but seems to just have a 10k resistor dropping off the same rail that V+ comes in on. It also says STBY pin threshold voltage is 1.5v. Assuming a 14v rail, a 10k resistor in series would drop the voltage to something minute wouldn't it? Anyways, I don't think this is the problem as I used a current limited seperate PSU to the same effect just to narrow the ridiculously long list of variables.
-Signal and power ground. Think I got this figured out at the basic level - two separate grounds that are tied together - but if there's something new people blow it on regarding this, I'm all ears. I'll say that my layout is probably the polar opposite of ideal
Onto the LM1875T...hoping, or at least lying to myself so I wouldn't be so dissapointed with the outcome with the TDA, that a botched early attempt just damaged the IC, I came to the conclusion that the 10% THD of the TDA and great on-paper specs of the Texas Instrument chip meant I should pursue that. Initially tried to set it up single rail, no luck at all. Put it away for a few days and decided, what, with all my success thus far, mine as well tack on learning about split supply (and bastardizing and horrendously rewiring an EI core transformer). This chip I at least put on some perfboard, and for the most part followed the layout guidelines in the datasheet using the components enumerated here: https://www.instructables.com/id/Tales-From-the-Chip-LM1875-Audio-Amplifier/
I was actually feeling relatively optomistic about at least getting some noise this time upon hookup but with a healthy dose of realism at the ready. Let's just say if anyone saw my transformer, they would think it was something someone in north korea cobbled together. Actually, that's giving it too much credit. I'm amazed there wasn't a short given it was reused magnet wire pulled from another transformer and I got a little rough with her a few times...Anyways, the primary is slightly unevenly distributed around core and then over that, the secondary is shockingly haphazard. Told myself just wanted a proof of concept and I"d do a proper one later. The turns are uneven resulting in a ~15v RMS side and a ~13 volt on the other end. The last probably relevant piece of info, an open circuit powering of the thing results in a 2 amp draw on the primary. This seems sketchy....
Anyways, from the secondary i hooked up the two rails to a bridge rectifier IC. Am I to connect the "ground" lead (center tap I think?) to the ground input of the LM1875 circuit? Can this even work with rails that are offset voltages like this. I'll point out a final mistake (I will say, I have learned a lot during all of this, outcome be damned) to bring her home...For some reason I was going forward with the idea that AC would go directly to the chip and therefore didn't take into account the increased voltage at the outputs of the rectifier...Let's just say that I am very close to the absolute ratings....
To summarize: guess I just have a few random questions that probably aren't even the right ones to be asking. Any suggestions for common beginner mistakes to probe (pun intended) or resources that break all this audio stuff down like I'm a five year old are also welcome. Thanks in advance.
bson:
--- Quote from: rectumfried on April 02, 2019, 01:35:21 am ---
-The datasheet talks about how the standby pin needs an external 5ma limited power supply but seems to just have a 10k resistor dropping off the same rail that V+ comes in on. It also says STBY pin threshold voltage is 1.5v. Assuming a 14v rail, a 10k resistor in series would drop the voltage to something minute wouldn't it?
--- End quote ---
It says the max stand-by pin current is 100µA. 14-1.5 = 12.5V, so a 10k resistor limits it to 1.25mA, which is an order of magnitude more than the required max. You can easily check the pin current by measuring the voltage across the resistor and dividing it by 10k.
What capacitors did you use on the inputs? Make sure to avoid polarized electrolytics for those - they will be reverse biased half the time and depending on cap used may not work at all. The caps with one white and one black plate are okay to have polarized, with the white plate indicating the positive side. Two black plates means unpolarized.
bson:
BTW, I'd strongly suggest getting the first working before moving on. If it's not working it's because you misunderstood something fundamental, and it's better to understand and sort that out before moving on and making the same errors elsewhere...
rectumfried:
I tried 150,000k resistance, no discernible difference. Still only the one channel works. For the non polarized caps, I used ceramic caps rated for 50v and for the polarized I used various electrolytics, rated at adequate voltage. I don't think it has anything to do with the standby pin as I am able to make the thing go into what seems to me, is a standby mode. I.E playing music, disconnect and a few seconds later, music ceases with the ohmic connection from the standby pin to +V severed. I'm still going with my gut. I think I fried it in the beginning. Doesn't make sense to me that only one channel would work like that and I am fairly certain I have it hooked up properly.
Navigation
[0] Message Index
Go to full version